ASTRONOMY and ASTROPHYSICS Evolution of the Colour-Magnitude Relation of Early-Type Galaxies in Distant Clusters
Total Page:16
File Type:pdf, Size:1020Kb
Astron. Astrophys. 334, 99–109 (1998) ASTRONOMY AND ASTROPHYSICS Evolution of the colour-magnitude relation of early-type galaxies in distant clusters Tadayuki Kodama1,2, Nobuo Arimoto2,4, Amy J. Barger1,3, and Alfonso Aragon-Salamanca´ 1 1 Institute of Astronomy, University of Cambridge, Madingley Road, Cambridge CB3 0HA, UK ([email protected]) 2 Institute of Astronomy, University of Tokyo, Mitaka, Tokyo 181, Japan 3 Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, Hawaii 96822, USA 4 Observatoire de Paris, Section de Meudon, DAEC, F-92195 Meudon Principle Cedex, France Received 21 November 1997 / Accepted 17 February 1998 Abstract. We present a thorough quantitative analysis of the models of elliptical galaxies with the C-M diagrams for early- evolution of the colour-magnitude relation for early-type galax- type galaxies in two distant clusters, Abell 2390 at z =0.228 ies in 17 distant clusters with redshifts 0.31 <z<1.27 using and Abell 851 at z =0.407, and showed that the origin of the the Kodama & Arimoto (1997) evolutionary model for ellipti- C-M relation was primarily due to the mean stellar metallicity of cal galaxies. The model is calibrated to reproduce the colour- the early-types being a function of total magnitude. In this paper magnitude relation for Coma ellipticals at z ∼ 0 and gives the we present a more thorough study of the evolution of the C-M evolution of the slope and zero-point as a function of redshift. We relation by using 17 clusters at cosmological distances to detect find no significant differences between the colour-magnitude the colour evolution of cluster early-type galaxies directly and relations of the clusters in our sample. The slopes can be repro- then comparing the results with the models. This will enable us duced by a single model sequence in which all elliptical galaxies to put even stronger constraints on the formation of early-type are assumed to be equally old (the maximum age difference al- galaxies in clusters. lowed for the brightest 3 magnitudes is only 1 Gyr) and to have The photometric evolution of early-type galaxies in z<1 mean stellar metallicities which vary as a function of galaxy clusters was previously examined by several authors (e.g., Ellis luminosity. The zero-points of the colour-magnitude relations et al. 1985; Couch, Shanks & Pence 1985; Aragon-Salamanca´ et constrain the epoch of major star formation in early-type galax- al. 1993). Aragon-Salamanca´ et al. (1993) traced the evolution ies to zf > 2–4. This study provides two important constraints of the ‘red envelope’ from the optical-near-infrared colour dis- for any model of the formation of rich clusters: the uniformity of tribution of galaxies in 10 clusters with redshifts 0.5 <z<0.9 the ages of the stellar populations in the early-type galaxies and and concluded that the detected evolution was consistent with the universality of the metallicity sequence of these galaxies as the passive ageing of stellar populations formed before z ' 2. a function of galaxy mass. A potential problem with these studies is that the early-type galaxies had to be selected by their spectral energy distributions Key words: galaxies: active – galaxies: elliptical and lenticu- and colours, thereby increasing the possibility of contamina- lar, cD – galaxies: evolution – galaxies: formation – galaxies: tion from other galaxy types. The advent of the Hubble Space photometry – galaxies: stellar content Telescope (HST), with its spectacular high-resolution images, has made the morphological classification of distant galaxies possible. Furthermore, thanks to the HST, large ground-based telescopes, and the high quality of astronomical detectors, much 1. Introduction fainter limiting magnitudes can now be reached. An additional Visvanathan & Sandage (1977) first noted from their data on advantage of HST photometry comes from the reduction of 9 nearby clusters that the more luminous early-type galaxies the sky background at longer wavelengths compared to that tended to have redder colours. Bower, Lucey, & Ellis (1992; achieved with ground-based observations (e.g. the background hereafter BLE92) later studied this colour-magnitude (C-M) re- in space is 8 times fainter in the I-band), thereby reducing the lation in detail using high precision photometry of early-type photometric errors at faint limits significantly (Ellis et al. 1997; galaxies in the Virgo and Coma clusters and found very little hereafter E97). These improvements now make it possible to scatter about the mean C-M relation. Thus, there appeared to examine the evolution of early-type galaxies over a much wider be a marked homogeneity in the present-day early-type clus- luminosity and redshift range. ter galaxy population. Recently, Kodama & Arimoto (1997; The C-M relation is a powerful tool for quantifying the hereafter KA97) compared the predictions of their evolutionary colour evolution of early-type galaxies as a function of redshift since it can be observed up to very high redshifts. Dickinson Send offprint requests to: T. Kodama (1996) showed that the C-M relation is already recognizable at 100 T. Kodama et al.: Evolution of the colour-magnitude relation of early-type galaxies in distant clusters z ' 1.2. The zero-point of the C-M relation at high redshift pro- the C-M relation and the zero-point, since it includes the effects vides direct information on the properties of early-type galaxies of metallicity. Moreover, since the model is calibrated with the in general. On the other hand, the slope of the C-M relation has empirical colours of ellipticals in Coma, it can be directly and information on differential properties of early-type galaxies as reliably compared to the observational data in distant clusters. a function of luminosity or galaxy size. In this paper, we investigate the evolution of the C-M relation Stanford, Eisenhardt, & Dickinson (1995; 1998, hereafter of 17 distant clusters using the KA97 evolutionary models. We SED98) presented C-M relations for morphologically selected have accumulated the photometric data from the literature, most early-type galaxies in 19 clusters out to z =0.9. When they of which were obtained with the HST and large, ground-based compared their optical-near-infrared C-M relations to that of telescopes (∼ 2–4 m). Although some of the HST data suffer Coma, they found no significant change in either the slope or from zero-point uncertainties due to the paucity of observed the scatter as a function of redshift. They did, however, observe standard stars (see below), their random photometric errors are a progressive blueing with redshift of the average colour in a very small down to ∼ 3 mag from the brightest end of the C- manner consistent with the passive evolution of an old stellar M relation (typically 0.01 − 0.07 mag). Thus, we are able to population formed in a burst at an early cosmic epoch. conduct a reliable analysis of, for example, the slope of the C-M E97 also analyzed the C-M relations of three distant clus- relation based on the relative photometry within each cluster. To ters at z ' 0.54 (Cl 0016+16, Cl 0054−27, Cl 0412−65), in constrain the formation epoch of early-type galaxies in clusters, this case using HST photometry. They found that the dispersion we also conduct a zero-point analysis using only those clusters about the relations was quite small even in such distant clus- which have been calibrated with ground-based photometry or ters, thereby requiring a high formation redshift for the stars those which are at sufficiently high redshift that the evolutionary in early-type galaxies, such as z ' 3. E97 also did not detect a changes are large enough to put some constraint on the formation significant change in the slopes of the relations in the rest-frame epoch, despite relatively large zero-point uncertainties. U −V when compared with Coma. The zero-points show mod- The cosmological parameters we have chosen to use −1 −1 est colour evolution in agreement with earlier studies (Ellis et throughout this paper are H0 =50km s Mpc (h ≡ al. 1985; Couch et al. 1985; Aragon-Salamanca´ et al. 1993). H0/100=0.5) and q0 =0.5, without a lambda term, un- All of these results support a formation picture of early-type less otherwise stated. The structure of this paper is as follows. galaxies in which the bulk of the star formation is completed In Sect. 2 we compile the observational data and define the ob- at high redshift with little star formation occurring in the re- served C-M relation for each cluster. In Sect. 3 we give a brief cent past. Thus, the C-M relation originates at high redshift as description of the evolutionary models of KA97. We compare a metallicity effect. the models and the observations in Sect. 4, and in Sect. 5 we Recently, however, Kauffmann & Charlot (1997) have ar- give a discussion of the results and our conclusions. gued that the C-M relation can alternatively be interpreted in the context of the hierarchical models of galaxy formation once 2. Data chemical enrichment is taken into account. In their analysis, 2.1. Sources the slope of the C-M relation is maintained because large ellip- ticals form primarily from large metal-rich progenitors. They Our dataset for this analysis consists of 16 clusters with red- find that it remains nearly constant up to z ' 1. Furthermore, shifts between z =0.31 and z =1.206 from various sources, they are able to explain the tightness of the C-M relation, de- all of which have been imaged with the Wide Field and Plane- spite frequent galaxy merging, by suggesting that we are biasing tary Camera 2 (WFPC2) on the HST.